See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/228499726

Alternative cost-effective preparation method

of polyaluminium chloride (PAC) coagulant
agent: Characterization and comparative
application for water/wastewater …

ARTICLE in DESALINATION · NOVEMBER 2009

Impact Factor: 3.96 · DOI: 10.1016/j.desal.2009.09.053

CITATIONS DOWNLOADS VIEWS

9 4,480 763

2 AUTHORS:

Anastasios Zouboulis Nikos Tzoupanos

Aristotle University of Thessaloniki Athens School of Fine Arts
235 PUBLICATIONS 4,863 CITATIONS 11 PUBLICATIONS 111 CITATIONS

SEE PROFILE SEE PROFILE

Available from: Anastasios Zouboulis

Retrieved on: 07 July 2015
 Desalination 250 (2010) 339–344

Contents lists available at ScienceDirect

Desalination
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / d e s a l

Alternative cost-effective preparation method of polyaluminium chloride (PAC)

coagulant agent: Characterization and comparative application for
water/wastewater treatment☆
A.I. Zouboulis ⁎, N. Tzoupanos
Division of Chemical Technology, Department of Chemistry, Aristotle University of Thessaloniki, GR-54124, Greece

a r t i c l e i n f o a b s t r a c t

Article history: An alternative preparation procedure for polyaluminium chloride (PAC), a common inorganic coagulant, is
Received 15 November 2007 presented in this paper. The proposed procedure is based on the use of granular aluminium metal, a common
Accepted 3 May 2008 by-product of several aluminium-processing secondary industries, instead of using Al2O3 (or of using a mineral
Available online 8 October 2009
containing Al2O3) which is commonly applied by most industrial producers, and exhibits important benefits
such as energy and time saving during the preparation procedure. Several efforts have been made to produce
Keywords:
Polyaluminium chloride
the pre-polymerized coagulant with properties, similar to the properties of a commercially available PAC
Coagulation solution, using appropriate, commonly found laboratory equipment and without the application of extreme
Water treatment conditions (in terms of temperature or pressure). The laboratory prepared PAClab was characterized mainly by
PAC application means of aluminium content, basicity, density and aluminium species distribution. Furthermore, the
coagulation performance of PAClab was evaluated for the treatment of contaminated tap water and of
wastewater samples and it was compared with the performance of a commercially available PAC solution.
Finally, the coagulation–flocculation kinetics was examined with the use of Photometric Dispersion Analyzer
instrument (PDA) and the floc growth rates between the two coagulants were compared. The results suggest
that PAC can be alternatively produced under mild conditions, resulting in a product with better properties
than the commercial PAC used for comparison. Additionally, if the preparation takes place in industrial scale, a
greater improvement is further feasible, due to the common equipment used, instead of using e.g. autoclaving.
© 2009 Elsevier B.V. All rights reserved.

1. Introduction
Several research efforts have been devoted to improving the
efficiency of coagulation–flocculation process, a basic and essential
treatment technique in water and wastewater treatment facilities. After
studying the chemistry and behavior of simple Al salts (i.e. of
conventional coagulants, such as alum/Al2(SO4)3), the way of the
improvement seemed to be the partial polymerization of these. The
result of these efforts was the production of a range of pre-polymerized
aluminium solutions, referred to as polyaluminium chlorides (PAC),
polyaluminium sulfates (PAS) and polyaluminium chloro-sulfates
(PACS) with variable degrees of polymerization. These products are
used extensively all over the world, especially during the last two
decades, with increasing demand. Their properties were intensively
examined and have proved to be more efficient in lower dosages and
☆ Presented at the 1st Conference on Environmental Management, Engineering,
Planning and Economics (CEMEPE), Skiathos, Greece, 24-28 June, 2007.
⁎ Corresponding author. Tel./fax: +30 2310 997794.
E-mail address: zoubouli@chem.auth.gr (A.I. Zouboulis).
higher pH, temperature and colloids range, than the conventional ones,
leading to cost and operative effective treatment [1,2].
Among the aforementioned products, polyaluminium chloride has
become one of the most effective coagulant agents in water and
wastewater treatment facilities with various applications, including
removal of colloids and suspended particles, organic matter, metal ions,
phosphates, toxic metals and color. The preparation of this reagent in
industrial scale generally involves the separate production of two
solutions that contain Al [3–5]. The first is called basic aluminium
chloride and is produced through the processing of an aluminium
containing raw material (e.g. Al2O3, or other Al containing mineral, such
as bauxite) under vigorous conditions (e.g. high temperature, pressure,
extended contact time) with concentrated acid (e.g. HCl, H3PO4 or an
appropriate mixture of them). The second solution is called sodium
aluminate (usually formulated as NaAlO2, Na2O·Al2O3, or Na2Al2O4 in
solid state) and can be produced by the dissolution of aluminium raw
material in a strong alkaline solution (e.g. NaOH, Na2CO3, NaHCO3, KOH
etc.). The production of PAC is accomplished by the introduction of
sodium aluminate solution into the basic aluminium chloride solution
under stirring, resulting in a product that can be described with the
generic formula Aln(OH)mCl3n−m (when HCl is used), where 0 b m b 3n.
The basic reactions that can describe the preparation of the two

0011-9164/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.desal.2009.09.053
340 A.I. Zouboulis, N. Tzoupanos / Desalination 250 (2010) 339–344
solutions are the following (with the use of hydrated alumina, HCl and
NaOH):
Al2 O3
3H2 O þ2nHCl→2AlCln ðOHÞ3−n þ2nH2 Oðbasic aluminium chlorideÞ
ð1Þ
Al2 O3 þ 2NaOH→2NaAlO2 þ H2 Oðsodium aluminateÞ: ð2Þ
Various PAC preparation methods exist, based on different starting
materials [4–9], as well as on certain modifications during the
preparation procedure, e.g. electrochemical dissolution of Al raw
material [10], without use of alkaline solution [6,11], or with the use of
stabilizing agents [12]. However, the procedures described in these
publications/patents are quite complicated, or the yields suggested
are questionable. The cost-effectiveness is a very important param-
eter, which designates (to a certain degree) the ability of a product to
be produced for commercial purposes. Considering the generic
preparation procedure for PAC solutions, it is understandable that
much energy is consumed particularly during the production of basic
aluminium solution. Moreover, extended contact time under in-
creased pressure is needed, which can be converted to production loss
for a continuous, full-scale operative facility. An alternative way for
the production of PAC should be desirable, resulting in a more
effectively preparation route, regarding time and cost.
In this paper, an alternative method for producing polyaluminium
chloride is presented, based on the use of granular Al, as aluminium
source. Particularly, the purpose was to synthesize a product with similar
properties to commercially available PAC (especially with the product
denoted as PAC-18). The laboratory prepared PAC was characterized by
means of Al% content, basicity, density and aluminium species distribution
(ferron). Its coagulation performance in contaminated tap water, as well
as in real wastewater samples was compared to the performance of
commercial PAC-18 sample and finally, the floc growth rates of the
respective coagulants were compared by means of the specific Photo-
metric Dispersion Analyzer (PDA) instrument.
2. Materials and methods
All chemical reagents used were analytically pure chemicals.
Deionized water with conductivity lower than 0.5μS/cm was used in
preparing all the solutions. For comparison purposes, commercially
available PAC-18 (17.15% Al2O3, basicity 40.1, density (d) 1.365 gcm−3)
was commercially obtained from PFI (Phosphates Fertilizers Industry SA),
Greece.
2.1. Synthesis of polyaluminium chloride
2.1.1. Preparation of solutions
Aluminium solution (about 9.5 g Al/100 mL) was prepared by
dissolving granular Al (Aldrich, ~40 mesh, or around 420 μm) in HCl
(min 37%). Granular Al was introduced slowly and in small portions to
preheated (65–70 °C) HCl solution, placed in a 500 mL flask under
continuous magnetic stirring. After the introduction of the first
portion of Al and the beginning of the (exothermic) reaction, further
heating was not needed.
Sodium aluminate solution was prepared in a similar way, but
without heating. A portion of NaOH solution (10–50%) was placed in a
500 mL flask on a magnetic stirrer. Under intensive mixing, granular
Al (10–15 g Al/100 mL) was slowly introduced and dissolved. Various
NaAlO2 solutions were prepared and used for PAC production. During
the preparations, small portions of acid and base solutions were also
added to replenish the respective loses, due to evaporation.
2.1.2. PAC preparation
A portion of aluminium solution was placed in a sealed flask on a
magnetic stirrer. With the use of a peristaltic pump, the appropriate
amount of sodium aluminate solution was slowly added (0.2 mL/min)
under vigorous stirring (1000 rpm). In the very dense sodium
aluminate solutions (50% NaOH, N12% Al), small portion of water
was added to lower the viscosity. The synthesis took place at various
temperatures, e.g. room temperature, 50 °C, 70 °C and 80 °C.
2.2. Characterization methods
Aluminium content (Al %w/w) and basicity (%) were determined,
according to AWWA Standard for PAC [13]. Density was measured
with a small 20 mL lykythos. Aluminium species distribution was
determined with the application of Al-ferron timed spectrophoto-
metric method, which is based on the different reaction time of
aluminium species with ferron reagent (8-hydroxy-7-iodoquinoline-
5-sulphonic acid) to form water soluble complexes at pH 5–5.2. These
complexes absorb light with maximum at 370 nm, hence absorbance
measurements at this wavelength allow the calculation of different
species of aluminium. Specifically, monomeric Al reacts almost
simultaneously (within 1 min) with ferron, while the intermediate
polymeric species of aluminium (mainly Al13, known as “Keggin”
structure) react slower, i.e. at 120 min. The larger and insoluble
polymeric structures (Alc, corresponding mainly to Al(OH)3) need
much more time to react, or it does not react at all. A UV–Vis
spectrophotometer was used for this purpose. The exact procedure
was a modification of the method of Parker and Bertsch [14],
developed by Zhou et al. [15].
2.3. Coagulation performance
The zeta-potential was measured by using a Laser Zee Meter 501,
the pH by using a Metrohm Herisau pH-Meter and the turbidity
measurements were performed on a HACH RATIO/XR Turbidimeter.
The absorbance at 254 nm (due to the Natural Organic Matter/NOM
content) was measured with a Schimadzu UV/Vis spectrophotometer,
using a 1 cm path length quartz cuvette. Residual aluminium
concentration was determined with eriochrome cyanine R method
[16]. The concentration of phosphates was determined with the
ascorbic acid method, according to APHA [16].
The coagulation performance of coagulants was evaluated in
contaminated tap (model) water sample, as well as in pre-treated
landfill leachates (aiming mainly to phosphates removal) with the use
of an appropriate jar-test apparatus (Aqualytic), having six paddles.
The contaminated (model) sample was made from tap water, clay
particles (kaolin suspension, commercially available, 10 mg/L) and
humic acid (Aldrich, 5 mg/L). As a flocculant aid anionic polyacryl-
amide (Magnaflock LT25, Ciba SC LTD) was used in concentrations
equal to the 1/10 of the respective concentration of the coagulants.
The jar test experimental conditions in the case of synthetic sample
were the following, based upon our previous experience: rapid mixing
at 160 rpm (or 200 s−1, expressed as velocity gradient units) for
2 min, slow mixing at 45 rpm for 10 min and sedimentation (settling)
time for 45 min. One minute after the addition of the coagulant (i.e.
during the rapid mixing stage) a 30 mL sample was withdrawn for z-
potential measurements. The flocculant aid was introduced just 15 s
before the slow mixing period. About 50 mL of sample was withdrawn
5 cm below the liquid surface for analytical determinations at the end
of these experiments. The dosages of the coagulants are expressed as
mg Al/L. In the case of leachates the coagulation conditions were the
following, based upon relevant preliminary experiments: rapid
mixing at 200 rpm for 3 min, slow mixing at 40 rpm for 30 min and
sedimentation for 45 min.
For the examination of coagulation–flocculation kinetics a test
sample of 1.5 L containing 5 mg/L clay particles and 5 mg/L humic acid
was placed in a 2 L baker and stirred with the paddle of a jar-test
apparatus. A peristaltic pump, placed after the PDA apparatus,
provided the necessary flow rate (30 ml/min), whereas the flowing
 A.I. Zouboulis, N. Tzoupanos / Desalination 250 (2010) 339–344 341

suspension was illuminated by a narrow light beam (of 850 nm As it can be observed from Table 1, the desired aluminium content
wavelength). All experiments were conducted at room temperature, (9% w/w) was not initially accomplished. The highest Al content was
without the addition of flocculant aid, to prevent the excessive growth found to be 7.3% w/w (PAC-G) and it was achieved with the sodium
of flocs and the blockage of connecting tubing (of diameter 3 mm). aluminate solution, prepared with 40% NaOH and 11% Al under room
The impact of velocity gradient during the rapid mixing stage (2 min temperature. For the NaAlO2 solutions having weaker basic character
duration) in floc growth was studied for the selected coagulants by (i.e. with NaOH content 10–30%), the amount of aluminium that could
monitoring the ratio (rms/dc), or Flocculation Index (FI) values. The FI be dissolved was restricted and the Al content of PAC was less than 7%
value is strongly correlated with the respective floc size and always in all cases (e.g. PAC-H). With the denser solutions (i.e. containing
increases as flocs grow larger [17]. The concentration of coagulants NaOH 50%) the addition of water in order to lower the viscosity,
was 2 mg/L. resulted in the decrease of Al content of produced PAC samples (PAC-
C). The amount of water needed was found to increase with the
increasing of Al content, while the use of sodium aluminate solutions
3. Results and discussion having 50% NaOH and higher (initial) Al content (e.g. 14%) was not
found to increase further the aluminium content in the final PAC
3.1. Synthesis of PAC: determination of the optimum preparation conditions product (PAC-D). Furthermore, the NaAlO2 solutions prepared with
higher NaOH and Al initial content were found to be rather unstable,
The purpose of this study, as aforementioned, was the preparation as a few days after their preparation they turned into gels.
of a polymerized inorganic coagulant/product with similar properties The temperature seems to exhibit a negative effect on the main
to the properties of commercially available PAC-18, but under mild physico-chemical properties of the final product. It was observed that
conditions, therefore with higher economic feasibility. More specif- during the synthesis of PAC, a white precipitate was formed
ically, the final product should contain at least 9.0% w/w Al, should (corresponding to Al(OH)3). With the increase in applied temperature
have basicity of around 40% and exhibit at least equal coagulation during the preparation, the rate of precipitate formation was also
performance with PAC-18. increased, resulting in the decrease of aluminium content, as seen in
The appropriate sodium aluminate solution should have sufficient the case of PAC-A (prepared under 80 °C), when compared to PAC-D
basic character for the subsequent neutralization of aluminium (prepared under room temperature). Additionally, the % basicity and
solution (polymerization), and at the same time enough Al content the % Al13 content for the PAC-A sample are the lowest between all
in order to increase the Al content of final product. The achievement of examined PAC solutions (28 and 16%, respectively). The same
high Al content (9%) could be problematic with the usual laboratory observations can be made in the case of applying lower than 80 °C
equipment and the initial efforts were focused on this issue. Various temperatures, e.g. PAC-F (prepared under 70 °C), as compared to PAC-
sodium aluminate solutions were used with different contents of G (prepared under room temperature). Hence, it can be concluded
NaOH (e.g. 10–50% NaOH) and aluminium (e.g. 10–15%). Further- that the increase in temperature during the synthesis of PAC results to
more, the synthesis took place at different temperatures (room temp., the deterioration of all examined PAC properties.
50 °C, 70 °C and 80 °C), to study the influence of temperature in the From the aforementioned experiments it is suggested that the
properties of the final product. The best results, regarding the most suitable composition of NaAlO2 is 40% NaOH and 11% Al (PAC-G)
composition of NaAlO2, were achieved with NaOH content N30% and and that preparation should take place under room temperature. In
Al content ≥11%. In Table 1 the major properties of selected PAC order to produce PAC with higher aluminium content (9% w/w), it is
samples produced with different NaAlO2 solutions, are presented. suggested that the restricted solubility of granular Al in HCl, due to
These include the % w/w Al content, the % basicity and the Al13 precipitate formation, should be enhanced. For this purpose, after the
content, determined through ferron assay. The Al13 (AlO4Al12(OH)24 addition of granular Al in HCl, the warm solution was placed in a
(H2O)126+,7+) with ε-Keggin structure is one among the many plastic bottle, sealed and kept in a furnace for 72 h under 90–95 °C.
possible PAC poly-nuclear compounds, maybe the most important), The increase in contact time under the higher temperature resulted in
being transformed continuously from one form to another [18]. Al13 is the increase in Al solubility and using this Al solution and the NaAlO2
claimed to be the most stable aluminium species in a partially solution, prepared from 40% NaOH and 11% Al, the preparation of PAC
neutralized aluminium solution [19] and the improved coagulation with 9% Al was finally achieved under room temperature. In Table 2
properties of PAC are thought to be due to its existence, i.e. increased the properties of laboratory prepared PAClab, of PAC-18 and the
charge neutralization capability (coagulation) and increased molec- specifications of PAC products, according to AWWA are given. PAClab
ular size and aggregation ability (flocculation). Further on, the seems to exhibit better properties than the commercial PAC-18
decrease of monomeric Al in favor of polymeric Al13 specie, as well product, higher basicity (44%), as well as higher Al13 content (29%).
as of the other polymeric compounds eliminates the hydrolysis The calculation of sodium aluminate amount for the neutralization
reactions and therefore, results in a minor impact of the resulting pH of the solution was based on the theoretical OH/Al molar ratio of about
value after treatment. 1.2. Certain efforts have been made by using higher basicity (e.g. OH/
Al 1.5), with increasing the amount of NaAlO2. It was observed that
Table 1 increasing the amount of sodium aluminate results in a slight increase
Properties of selected PAC solutions prepared under different experimental conditions. in basicity (45.1%), but at the same time the precipitate formation
PAC Composition of Theoretical Temperature PAC properties increased and in the final product PAC-B the Al content (6%) was less,
sample NaAlO2 solution OH/Al during
Al Basicity Al13
molar synthesis
a (%w/w) (%) (%)
ratio Table 2
A 50% NaOH + 14% Al 1.2 80 °C 5.5 28.3 18 Properties of laboratory prepared PAClab, PAC-18 and specifications for PAC products
B 50% NaOH + 10% Al 1.5 Room temp. 6 45.1 31 according to AWWA [13].
C 50% NaOH + 10% Al 1.2 Room temp. 7 35 21
PAC Al Basicity Al13 Density
D 50% NaOH + 14% Al 1.2 Room temp. 7 42.7 26
(% w/w) (%) (%) (g/cm3)
E 40% NaOH + 12% Al 1.2 Room temp. 7.1 39 23
F 40% NaOH + 11% Al 1.2 70 °C 6.8 34 21 AWWA 2.5–13 10–83 N/Aa N/A
G 40% NaOH + 11% Al 1.2 Room temp. 7.3 38.3 23 PAC-18 9.08 40.1 25 1.365
H 30% NaOH + 10% Al 1.2 Room temp. 6.9 36 23 PAClab 9.05 44 29 1.361
a a
Basicity = ([OH/Al]/3) ⁎ 100, according to [2]. Not available.
342 A.I. Zouboulis, N. Tzoupanos / Desalination 250 (2010) 339–344
than in the respective PAC-C (7% Al) prepared with the same NaAlO2
solution, but with lower OH/Al ratio (1.2). Therefore, in this case and
in order to induce the Al solubilization closed reactors (autoclaves)
have to be applied.
3.2. Coagulation performance of laboratory prepared PAC and
comparison with commercially available PAC-18
3.2.1. Simulated surface water treatment
The coagulation behavior of PAClab was evaluated in the (model)
clay–humic sample, simulating contaminated surface water (see the
previous §2.3) and compared with the coagulation behavior of PAC-
18. The initial turbidity of the sample was 16.1 NTU, the initial
absorbance at 254 nm 0.126, the pH value 7.65 and the applied
coagulant doses were 1–5 mg Al/L. Fig. 1 demonstrates the coagula-
tion performance of two coagulants. Removal of turbidity (Fig. 1a) and
of absorbance at 254 nm (Fig. 1b), residual aluminium concentration
(Fig. 1c) and z-potential variation of the suspensions after the
addition of coagulant (Fig. 1d) are presented.
PAClab seems to exhibit better coagulation performance than PAC-
18, especially in very low coagulant dose (e.g. 1 mg Al/L). Regarding
turbidity and absorbance reduction, similar behavior was observed,
but for the coagulant dose of 2 mg Al/L, PAC-18 was found to be
slightly more efficient. In all other concentrations PAClab is more
efficient. Regarding residual aluminium concentration, PAClab exhibits
also better performance. The Al concentration that remains in the
suspensions after the coagulation is lower for the case of PAClab
samples and for almost all the coagulants concentrations, although for
Fig. 1. Coagulation performance of PAC-18 and of laboratory prepared PAC (PAClab) in contaminated tap (model) water sample; a) residual turbidity (initial turbidity 16.5 NTU),
b) absorbance at 254 nm (initial absorbance 0.125), c) residual aluminium concentrations, d) z-potential measurements, with respect to coagulant dose.
the addition of 5 mg Al/L coagulant, PAClab and PAC18 show rather
similar performance.
The superiority of PAClab can be attributed to the higher basicity of
this product, and consequently to the higher Al13 content, than the
respective values of PAC-18. From the z-potential measurements it is
also quite obvious that the PAClab has a greater impact in the charge of
particulates. This is an indication of higher basicity and higher
coagulant charge, resulting in more effective charge neutralization,
colloids destabilization and consequently, in better coagulation
performance.
3.2.2. Wastewater (tertiary) treatment
The coagulation efficiency of PAC-18 and PAClab was also tested for
the tertiary treatment of biologically pretreated landfill leachates, and
for coagulant doses of 10–100 mg Al/L. The initial characteristics of
wastewater sample were: turbidity 2.1 NTU, absorbance at 254 nm 2.2
and phosphates concentration 32.9 mg/L. Fig. 2 illustrates the
coagulation performance of the tested coagulants.
Regarding turbidity removal (Fig. 2a) both coagulants seem to be
quite efficient, as a dose of 10 mg Al/L for both coagulants is enough to
lower turbidity values below 1 NTU. PAClab seems to exhibit a slightly
better performance, especially in doses of 50–100 mg Al/L. Regarding
absorbance reduction (Fig. 2b) both coagulants are less efficient,
however PAClab is more efficient in almost all cases, except for
coagulant dose of 10 mg Al/L.
Regarding the phosphates removal (Fig. 2c), it is clear that PAClab
exhibits better performance, than PAC-18. For all coagulant doses the
phosphates concentration of the treated sample was lower for PAClab,
with only the exception of 10 mg Al/L dose. Moreover, the
 A.I. Zouboulis, N. Tzoupanos / Desalination 250 (2010) 339–344 343

Fig. 2. Coagulation performance of PAC-18 and laboratory prepared PAC (PAC lab) for the treatment of real wastewater sample; a) residual turbidity (initial turbidity 2.1 NTU),
b) absorbance at 254 nm (initial absorbance 2.2), c) phosphates concentration (initial conc. 32.9 mg/L), d) phosphates (% percent removal), with respect to coagulant dose.

concentration of PAClab needed for limiting the concentration of longer for the case of PAC-18, especially for the velocity gradients 241
phosphates under 1 mg/L (i.e. the respective legislation limit for and 282 s− 1. Moreover, the variation of ratio values with the velocity
sensitive areas) is about 50–60 mg Al/L, whereas with the case of PAC- gradients has similar profile for both coagulants, i.e. greater Ratio
18 the respective concentration needed is more than 60 mg Al/L. For values were noticed for 200 s− 1, whereas these values were
the addition of 70 mg Al/L (or even higher) the concentration of decreased, when increasing the stirring speed. It seems that by
phosphates is minimized for both coagulants, with slightly lower increasing the shear stress results in floc breakage for both coagulants.
values found for the case of PAClab addition (Fig. 2d). A significant difference between the two coagulants can be
observed by notifying the rate of Ratio values increment at the
3.2.3. Coagulation–flocculation kinetics beginning of flocculation stage. This increment was more intense for
In Fig. 3 the respective Ratio values (as given by the instrument) the case of PAClab, indicating that with PAClab the flocculation occurs
vs. time for the two coagulants are presented. Three different velocity
gradients were applied during the rapid mixing stage, i.e. 160, 180, or
200 rpm (or 200 s− 1, 241 s− 1 and 282 s− 1), expressed as velocity
gradient units respectively. The monitoring of Ratio (or FI) values
during the PDA experiments provides useful information, concerning
the potential of coagulants to form flocs and allows certain
comparisons to be made between the different coagulants. Specifi-
cally, it allows the relative comparison of the extent of aggregation, as
higher ratio values indicate bigger floc size. Furthermore, the duration
of lag phase, i.e. the fast mixing period where coagulation occurs and
the ratio remains relatively low and invariable, represents another
comparison parameter. Finally, the rate of floc growth immediately
after coagulation and at the beginning of flocculation stage, where the
FI values increase suddenly, can also be used for comparison reasons.
From Fig. 3 it can be deduced that with the addition of PAClab the
respective Ratio reaches higher values, than with the case of PAC-18,
when comparing them for the same stirring velocities. It can be
supposed that with the laboratory prepared coagulant bigger flocs are Fig. 3. Coagulation kinetics monitored by the use of Photometric Dispersion Analyzer
formed, resulting probably in a more effective flocculation and for the addition of PAClab and of PAC-18 samples; the concentration of coagulants was
sedimentation. Regarding the duration of lag phase, it lasts slightly 2 mg/L.
344 A.I. Zouboulis, N. Tzoupanos / Desalination 250 (2010) 339–344
Table 3
Fitting equations and correlation coefficients of the curves from PDA experiments.
Coagulant Velocity gradient Fitting curve Correlation coefficient (R2)
(s− 1)
PAClab 200 r = 0.0097t − 0.2573 0.9862
241 r = 0.0094t − 0.1793 0.989
282 r = 0.0078t − 0.1372 0.994
PAC18 200 r = 0.009t − 0.1032 0.9919
241 r = 0.0075t − 0.1782 0.9909
282 r = 0.0073t − 0.1246 0.9856
quicker, than with the addition of PAC-18. In Table 3 the fitting
equations of the curves from Fig. 3 and for the time period 100–200 s
(i.e. before they reach their maxima) are presented. From the
respective correlations coefficients it seems that all these equations
are sufficiently well fitted with linear equations (as the correlation
coefficient R2 N 0.98). Furthermore, the slopes of fitting equations for
the case of PAClab sample are greater than for PAC-18 (for the same
stirring conditions), indicating the more rapid progress of particle
aggregation and of floc size increment.
4. Conclusions
With the use of solid granular aluminium, the application of
common laboratory equipment and without the application of
extreme conditions (in terms of temperature or pressure), the
production of dense polyaluminium chloride solution (PAClab) for
commercial purposes was accomplished. PAClab exhibits improved
properties, in comparison with the commercially available PAC-18,
having the same Al content (~9% w/w), but slightly higher basicity
and Al13 content. Due to these improved properties, PAClab shows
superior coagulation performance than PAC-18, when applied for the
treatment of simulated (model) contaminated natural water, or for
real wastewater (biologically pre-treated leachate) samples. Further-
more, it has been shown that PAClab enhances the floc size and growth
and accelerates the process of particle aggregation.
The aforementioned procedure has the benefits of energy and time
saving during the manufacture of PAC, resulting in lower operating
cost, which may compensate for the higher cost of metallic Al, used as
raw material. Considering that various aluminium by-products (e.g.
those found in wastes from recycling refreshment cans) could be used
for PAC production after the appropriate treatment, the cost
effectiveness option could be further improved. Finally, if the
described procedure takes place in industrial scale, further improve-
ment of the final product and refining of the method may be feasible,
due to the appropriate equipment used.
Acknowledgements
Thanks are due to the Greek Secretariat for Research and
Technology (Ministry of Development) for the funding of this
research, which is part of the PhD. thesis of Mr. N. Tzoupanos,
through the PENED program, as well as to European Union for the
partial co-founding of this program.
References
[1] S. Sinha, Y. Yoon, G. Amy, J. Yoon, Determining the effectiveness of conventional
and alternative coagulants through effective characterization schemes, Chemo-
sphere 57 (2004) 1115–1122.
[2] J.C. Crittenden, R.R. Trussel, D.W. Hand, K.J. Howe, G. Tchobanoglous (Eds.),
Coagulation, mixing and flocculation, in Water Treatment: Principles and Design,
2nd edition, John Wiley & Sons, New Jersey, 2005, pp. 664–691.
[3] J.M. Dulko. Polyaluminum chlorides and polyaluminum chlorosulfates methods and
compositions. (1999) United States Patent 5985234.
[4] M. Kvant and K. Stendahl. Process for the preparation of polyaluminum compounds.
(1993) United States Patent 5182094.
[5] P. Dufour. Process for the preparation of basic polyaluminum chlorosulphates and
applications thereof. (1999) United States Patent 5879651.
[6] O. Carlsson, 1987. Process for producing a flocculating agent. (1987) United States
Patent 4654201.
[7] L. K. Gunnarsson, R. Nilsson. Method for producing basic aluminum sulphate (III).
(1986) United States Patent 4563342.
[8] B. Pozzoli. Aluminium polychlorosulphates, process for their preparation and use
thereof. (2003) United States Patent 6548037.
[9] A.B. Gancy, C. Wamser. Amorphous polyaluminum sulfate compositions. (1980)
United States Patent 4238347.
[10] G. Kudermann, K.-H. Blaufuss. Method for increasing the molar ratio of aluminum
to chlorine in polyaluminum chloride solutions. (1993) United States Patent
5254224.
[11] G. Haake, G. Geiler, F. Haupt. Process for preparing sulfate-containing basic
solutions of polyaluminumchloride. (2001) United States Patent 6241958.
[12] G. Lindahl. Stable solutions of basic aluminium sulphate containing polynucleate
aluminium hydroxide sulphate complexes. (1985) United States Patent 4536384.
[13] AWWA standard for liquid polyaluminum chloride, American Water Works
Association (AWWA), Denver, Colorado USA, 1999.
[14] D.R. Parker, P.M. Bertsch, Identification and quantification of the ‘Al13’ trideca-
meric polycation using ferron, Environ. Sci. Technol. 26 (1992) 908–914.
[15] W. Zhou, B. Gao, Q. Yue, L. Liu, Y. Wang, Al-Ferron kinetics and quantitative
calculation of Al(III) species in polyaluminium chloride coagulants, Colloids Surf.,
A Physicochem. Eng. Asp. 278 (2006) 235–240.
[16] L. Clesceri, A. Greenberg, R. Trussell (Eds.), Standard methods for the examination
of water and wastewater, 17th Ed, APHA-AWWA-WEF, Washington DC, 1989.
[17] M.A. Yukselen, J. Gregory, The reversibility of floc breakage, Int. J. Miner. Process.
73 (2004) 251–259.
[18] S. Bi, C. Wang, Q. Cao, C. Zhang, Studies on the mechanism of hydrolysis and
polymerisation of aluminium salts in aqueous solution: correlations between the
Core-links model and Cage-like keggin-Al13 model, Coord. Chem. Rev. 248 (2004)
441–455.
[19] J.-Q. Jiang, N.G.D. Graham, Pre-polymerised inorganic coagulants and phosphorus
removal by coagulation — a review, Water S.A. 24 (1998) 237–244.